Heavy Oil Field Development Study

Context: The client was launching, together with a National Oil Company (NOC), a joint study of a difficult, heavy oil field. The client needed the study to show-case its expertise and demonstrate to the NOC the benefits that might arise if the client were given acreage to operate. Unfortunately, the work-load on the client’s existing acreage had just gone up. Where would they find at short notice highly-expert staff to assist on the project?

SERAFIM Consultancy Contribution: SERAFIM supplied a reservoir engineer and petrophysicist to be part of the client’s multi-discipline team and work in conjunction with the client and NOC reservoir engineers. The SERAFIM engineer helped with log analysis, well-test analysis, reservoir simulation and training and support of the NOC engineers.

Benefits: The SERAFIM reservoir engineer helped resolve the uncertainties about asphaltene behaviour and the discrepancies between well-test behaviour and laboratory PVT measurements. The NOC staff expressed themselves as very pleased with the training and support.

Free gas:liquid Ratio*

Context: our client, then a small producer with operations in Western Africa needed to model thin oil rim fields, with extensive gas coning. The bigger fields had reservoir simulation models, but the smaller and older fields did not. For investment and planning reasons, our client needed to forecast overall gas and oil profiles using the basic information on:

• Past production
• Gas coning
• The Geology of the reservoirs

SERAFIM Consultancy contribution: We derived a new algorithm to forecast associated gas production. The algorithm was based on the observation that, once the gas cones reach the production wells, the gas-liquid contact tends to reach an equilibrium position, with an associated equilibrium free-gas:liquid ratio that can be calculated from the gas and liquid volumes in place. The new algorithm was implemented in a software routine as part of the client’s decline analysis work-flow.

Benefits: The resultant gas production forecasts have proved robust and suitable for both medium and long-term production forecasting. The insight into the fundamental mechanisms of oil-rim gas coning has helped determine the optimal choice of well cut-back criteria.

C-Curves†

Context: In a combined reservoir/facilities/economics model of a heavy oil field, decline curves were being used to interpolate between the results of a small number of simulator runs. However, neither exponential nor hyperbolic curves gave a good match to the simulator profiles, whose decline rates diminished with time and then stabilized.

SERAFIM Consultancy Contribution: The SERAFIM reservoir engineer derived a mathematical formulation more general than the hyperbolic curve and successfully used it to match the simulator profiles.

Benefits: The combined reservoir/facilities/economics model was successfully calibrated to the simulator runs and then used for multiple optimisations, sensitivities and Monte Carlo calculations that would have taken an impossibly long time to calculate if the simulator had been directly linked into the combined runs. The client was able to analyse the full range of field development options, taking into account the key uncertainties in STOIIP, reservoir quality, well deliverability and reliability.

† Details in SPE paper 9609

Production Allocation

Context: A major North Sea operator approached us to solve a production allocation problem. They had four wells producing to a subsea manifold, which in turn channels the multiphase flow through a 30km pipeline to a FPSO. The data they had consisted in downhole gauges readings of THP, THT, BHP and BHT for each well and the aggregate production measured at the FPSO. In theory, the data was sufficient to determine the individual well production rates, but the existing allocation calculation was giving incorrect results.

Allocation problems usually require strong engineering and mathematical skills.

SERAFIM Consultancy Contribution: SERAFIM examined all the aspects of the problem, from the reservoir to the details of the original tool measurements to the mathematics of the allocation calculation. The analysis showed that the allocation calculation was using an algorithm that was mathematically inappropriate for data with the level of noise seen in the gauge measurements, and that earlier attempts to solve the problem had been hindered by errors in data transfer. A software routine was written, implementing an alternative, more robust allocation calculation and this was applied to both current and historical production data.

Benefits: The new production allocation resolved a number of anomalies in apparent reservoir behaviour and has simplified the reservoir simulation history match, leading to a reservoir simulator model that can be used with more confidence to commit to planned new wells.